JEQ Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 9 January 2007
Published in J Environ Qual 36:101-109 (2007)
DOI: 10.2134/jeq2005.0415
© 2007 American Society of Agronomy, Crop Science Society of America, and Soil Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hiradate, S.
Right arrow Articles by Fujii, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hiradate, S.
Right arrow Articles by Fujii, Y.
Agricola
Right arrow Articles by Hiradate, S.
Right arrow Articles by Fujii, Y.
Related Collections
Right arrow Agricultural Pesticides
Right arrow Soil Organic Matter
Right arrow Soil Surface Chemistry
Right arrow Sorption/Exchange
Right arrow Volcanic Soils

TECHNICAL REPORTS

Organic Compounds in the Environment

Adsorption of 2,4-Dichlorophenoxyacetic Acid by an Andosol

Syuntaro Hiradatea,*, Akihiro Furubayashia, Natsuyo Uchidab and Yoshiharu Fujiia

a Biodiversity Division, National Institute for Agro-Environmental Sciences (NIAES), 3-1-3 Kan-nondai, Tsukuba, Ibaraki 305-8604, Japan
b Environmental Research Center, 3-1 Hanare, Tsukuba, Ibaraki 305-0857, Japan

* Corresponding author (hiradate{at}affrc.go.jp)

Received for publication November 2, 2005.

   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
To identify the important soil components involved in 2,4-dichlorophenoxyacetic acid (2,4-D) adsorption on Andosols, 2,4-D adsorption on a surface horizon of an Andosol was compared with that on hydrogen peroxide (H2O2)-treated (soil organic matter [SOM] was removed), acid-oxalate (OX)-treated (active metal hydroxides and SOM were removed), and dithionite-citrate-bicarbonate (DCB)-treated (free and active metal [hydr]oxides and SOM were removed) soil samples at equilibrium pHs ranging from 4 to 8. Although the untreated soil contained a large amount of organic C (71.9 g kg–1), removal of SOM had little effect on 2,4-D adsorption. Active surface hydroxyls, which were attached to the active and free metal (hydr)oxides and metal SOM complexes, were identified as the most important soil functional group for 2,4-D adsorption. The dominant mechanism of the 2,4-D adsorption was a ligand exchange reaction in which the carboxylic group of 2,4-D displaced the active surface hydroxyl associated with metals and formed a strong coordination bond between the 2,4-D molecule and soil solid phase. The ligand exchange reaction reasonably accounted for the selective adsorption of 2,4-D over Cl–, competitive adsorption of phosphate over 2,4-D, reduction in plant-growth-inhibitory activity of soil-adsorbed 2,4-D, and the high 2,4-D adsorption ability of Andosols. Although a humic acid purified from the soil did not adsorb 2,4-D, the presence of the humic acid increased 2,4-D adsorption on Al and Fe, probably by inhibiting the hydrolysis and polymerization of Al and Fe resulting in the preservation of available adsorption sites on these metals. The adsorption behavior of 2,4-D on soils could be a good index for predicting the adsorption behavior of other organic acids in soils.

Abbreviations: AAS, atomic absorption spectrometry • 2,4-D, 2,4-dichlorophenoxyacetic acid • DCB, dithionite-citrate-bicarbonate • H2O2, hydrogen peroxide • OX, oxalate • SOM, soil organic matter


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
THE SELECTIVE and systemic herbicide 2,4-dichlorophenoxyacetic acid (2,4-D, Fig. 1) is used to control broadleaf weeds after emergence. Although 2,4-D is one of the oldest chemical herbicides in the world (created in the early 1940s), it is still widely used in agricultural fields because of its excellent selectivity between the broadleaf weeds and graminaceous crops and its reasonable cost. The strong auxinic action of 2,4-D is believed to persist in plants without significant biodegradation and to disturb hormonal balance, resulting in plant death.


Figure 1
View larger version (7K):
[in this window]
[in a new window]

  		Fig. 1. Chemical structure of 2,4-dichlorophenoxyacetic acid (2,4-D).

 
While 2,4-D has been proven useful as a chemical herbicide, its environmental contamination has become a concern because it is suspected to exhibit endocrine-disrupting activities (Rawlings et al., 1998; Short and Colborn, 1999). Contamination by 2,4-D has been reported in groundwater aquifers and surface water (Leonard, 1990; Hall et al., 1993; Wood and Anthony, 1997; Balinova and Mondesky, 1999). This has led researchers to study the adsorption mechanisms of 2,4-D on soils.

The 2,4-D molecule has three possible adsorption mechanisms on soils: (i) electrostatic ion exchange reaction between a positively charged, active surface hydroxyl on metal (hydr)oxides and the negatively charged carboxylate group on the dissociated 2,4-D molecule (Fig. 2A), (ii) ligand exchange reaction between the active surface hydroxyl on metal (hydr)oxides and the carboxylic group of the 2,4-D molecule (Fig. 2B and C), and (iii) hydrophobic interactions between hydrophobic constituents of soils (e.g., humic substances) and hydrophobic moieties of the 2,4-D molecule (e.g., aromatic and CH2 moieties). At present, however, a consensus about the mechanisms of 2,4-D adsorption by soils has not been reached. For example, Kavanagh et al. (1977) reported that goethite (iron oxyhydroxide) adsorbed 2,4-D primarily by an ion exchange reaction, although the amount of 2,4-D adsorption was considerably greater than would be expected purely on the basis of a simple electrostatic interaction. This result was attributed to the existence of favorable hydrophobic interactions (van der Waals interactions) between the adsorbed 2,4-D molecules themselves (that is, condensation). Madrid and Diaz-Barrientos (1991) also emphasized the importance of an ion exchange reaction to the adsorption of 2,4-D on lepidocrocite (iron oxyhydroxide). Koskinen and Harper (1990), however, questioned the predominant role of ion exchange reactions on 2,4-D adsorption on soils because layer silicate clays and soil organic matter (SOM) are generally either uncharged or negatively charged. Watson et al. (1973) proposed that the ligand exchange reaction is the dominant 2,4-D adsorption mechanism on goethite. Hydrophobic interactions between 2,4-D and SOM have also been proposed to contribute to 2,4-D adsorption by soils (Khan, 1973; Almendros, 1995; Benoit et al., 1996). In fact, 2,4-D adsorption on soils has frequently been evaluated on the basis of the soil sorption coefficient (KOC), defined as a distribution coefficient of a compound adsorbed onto the soil (Kd) divided by the organic carbon content of the soil (KOC = Kd/organic C content), which basically assumes partitioning of 2,4-D between SOM and the solution phase (OECD, 1981; von Oepen et al., 1991). Mallawatantri and Mulla (1992) questioned the efficiency of KOC values for estimating 2,4-D adsorption on soils because KOC values for 2,4-D depend on SOM contents. Any of these adsorption mechanisms could be possible depending on soil conditions (e.g., kind of major soil constituents, concentration of 2,4-D, equilibrium pH); but in any case, the primary adsorption mechanism at conditions typical of a given kind of soil should be clarified.


Figure 2
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 2. Scheme of 2,4-D adsorption on soils. A: 2,4-D adsorption on active surface hydroxyls of metal (hydr)oxides by an ion exchange reaction (electrostatic interaction). B: 2,4-D adsorption on metal (hydr)oxides by a ligand exchange reaction which replaces an active surface hydroxyl with a carboxyl group of 2,4-D molecule, forming a strong coordination bond. C: 2,4-D adsorption on metal-humate complexes by a ligand exchange reaction. M denotes metals such as Al and Fe.

 
Similarly, other organic acids can be adsorbed by both anion exchange and ligand exchange reactions because they have at least one carboxylic group. When an organic acid has a highly hydrophobic moiety in its structure, hydrophobic interactions with SOM could also occur. Recent research clearly indicates the important role of organic acids in plant–plant chemical communications (allelopathy), acquiring plant nutrition, and detoxifying Al (Buchanan et al., 2000). To be able to predict the behavior of these organic acids in soil environments, their adsorption mechanism should be clarified. Research on the adsorption mechanism of 2,4-D by soils could be a good reference in understanding the adsorption mechanism of other organic acids by soils because 2,4-D, like many other organic acids, has one carboxylic group (pKa 2.81, Bintein and Devillers, 1994) and a moderate hydrophobicity (octanol-water distribution coefficient KOW: 0.027 in alkaline condition, 29.23 in acidic condition, Wang et al., 1994).

In the present study, 2,4-D adsorption by soils was studied at the low 2,4-D concentrations likely to occur in soil environments (mostly <5.0 mg L–1). The standard application rate of 2,4-D on an agricultural field is 0.28 to 2.3 kg ha–1 (Tomlin, 2003), resulting in 0.56 to 4.6 mg L–1 of maximum 2,4-D concentration in paddy-flooded water if the flood depth is assumed to be 5 cm. To elucidate the major soil components causing 2,4-D adsorption, we compared 2,4-D adsorption by untreated soil material collected from a surface horizon of an Andosol with that by hydrogen peroxide (H2O2)-treated (SOM was removed), acid-oxalate (OX)-treated (active metal hydroxides and SOM were removed), and dithionite-citrate-bicarbonate (DCB)-treated (free and active metal (hydr)oxides and SOM were removed) soil samples. Adsorption of 2,4-D was monitored in the equilibrium pH range between 4 and 8 at constant electrolyte concentrations (0.01 or 0.1 M CaCl2), and the adsorption mechanism of 2,4-D was explored.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 
Soil Preparation
In the present study, samples collected from the surface horizon of a SOM-rich Andosol were used to study pH-dependent 2,4-D sorption characteristics. These soil samples were prepared as described by Hiradate and Uchida (2004), as follows. The soil sample was collected from the Ap horizon of an upland experimental field of the National Institute for Agro-Environmental Sciences, Tsukuba, Japan (Melanudand; USDA Soil Taxonomy; Soil Survey Staff, 1999; Silandic Andosol; WRB; FAO, 1998). The soil was air-dried, ground, and finely sieved (<0.5 mm) to prepare it for analysis (untreated UG5 soil). The untreated UG5 soil was put into a tall beaker and treated repeatedly with 6% H2O2 with boiling until SOM was completely decomposed, and washed once with 20 g L–1 Na2CO3 (without pH adjustment) to remove organic molecules adsorbed on the soil particles, three times with 1 M NaCl at pH 5, and three times with deionized water. The treated soil sample was freeze-dried and sieved (<0.5 mm, UG5-SOM). To remove the active metal hydroxides, the UG5-SOM was shaken with 0.2 M sodium oxalate-oxalic acid mixed solution (pH 3.0, soil sample/solution = 1:25, mass/volume) at 150 rpm in the dark for 4 h at 30°C. The treated soil sample was washed with 20 g L–1 Na2CO3 (without pH adjustment), 1 M NaCl, and deionized water, then freeze-dried and sieved, as described above (UG5-SOM-AM). To remove the free metal (hydr)oxides, 10 g of UG5-SOM-AM were suspended in a mixed solution of 100 mL of 0.3 M sodium citrate and 12.5 mL of 1 M sodium bicarbonate at 80°C, and 1 g of sodium dithionite was added. The mixture was incubated for 15 min with occasional shaking, and the supernatant was removed by centrifugation (ca. 300 x g). This procedure (DCB-treatment) was repeated twice. The treated soil sample was washed, freeze-dried, and sieved, as described above (UG5-SOM-FM).

Soil Analyses
Soil chemical analyses were also conducted as described by Hiradate and Uchida (2004), as follows. The determination of soil pH in H2O, KCl, and NaF was performed using a pH meter (HM-20E; TOA Electronics, Tokyo, Japan) with a glass electrode (GS-5017C, combination type for test tube, 8 mm {varphi}, TOA Electronics). Ten grams of air-dried soil were mixed with 25 mL of deionized water or 1 M KCl for the determination of the pH in H2O and KCl, respectively. The soil-solution mixture was allowed to stand for 12 h with occasional shaking, and the pH was measured in the suspension. To determine the pH in NaF, 1 g of air-dried soil was mixed with 50 mL of 1 M NaF (pH 7.0) and stirred continuously for 2 min, and then the pH of the suspension was measured. The contents of soil organic C and N were determined with a C/N analyzer (dry combustion method). Phosphate retention capacity was measured by the method reported by Blakemore et al. (1987). In this method, 5 g of air-dried soil (<2 mm) were equilibrated with 25 mL of a 1000 mg P L–1 solution (buffered at pH 4.6 by acetic acid) for 16 h at 20°C with shaking (50 rpm). After centrifugation, P concentration in the supernatant was determined by the vanadomolybdophosphoric acid method.

To determine the contents of pyrophosphate-extractable Al (Alpyr) and Fe (Fepyr), 1 g of soil sample was added to 100 mL of a 0.1 M sodium pyrophosphate solution and shaken for 16 h (van Reeuwijk, 1993). After the addition of 3 to 4 drops of 2 g L–1 Accofloc (N-100; Mitsui Chemical AquaPolymer, Tokyo, Japan), the soil-solution mixture was centrifuged for 10 min at 12 000 rpm (18 000 x g). The concentrations of Al and Fe in the supernatant were determined by atomic absorption spectrometry (AAS; Z-5010; Hitachi, Tokyo, Japan). To determine the contents of acid oxalate-extractable Al (AlOX), Fe (FeOX), and Si (SiOX), 1 g of soil sample was added to 50 mL of a 0.2 M ammonium oxalate-oxalic acid mixed solution (pH 3.0) and shaken for 4 h in the dark (van Reeuwijk, 1993). The concentrations of Al, Fe, and Si in the supernatant were determined by AAS. The content of dithionite-citrate-bicarbonate-extractable Fe (FeDCB) was determined by applying the method of Mehra and Jackson (1960) and by AAS analysis.

Adsorption of 2,4-Dichlorophenoxyacetic Acid by Soil Samples
A 2,4-D stock solution (489 µM, 108 mg L–1) was prepared by dissolving authentic 2,4-D (acid form; Wako Pure Chemical, Osaka, Japan) into deionized water (water solubility, 311 mg L–1 at pH 1, 20 031 mg L–1 at pH 5; Tomlin, 2003). A 250-mg portion of the soil sample (oven-dry basis) was placed in a glass centrifuge tube (12 mL). Deionized water (3.8 to 4.7 mL), 1 M CaCl2 (0.05 or 0.5 mL), 0.1 M HCl or NaOH (0 to 0.50 mL), and then 489 µM 2,4-D aqueous solution (0.232 mL) were added to the tube by using continuously adjustable air displacement pipettes (P-1000 and P-200, relative precision < ± 1% standard deviation, Gilson, WI, USA) to achieve a final volume of 5.0 mL, an equilibrium pH value between 4 and 8, a CaCl2 concentration of 0.01 or 0.1 M, and an initial 2,4-D concentration of 22.6 µM (5.0 mg L–1). The mixture of soil sample and 2,4-D solution was shaken at 120 rpm for 1 to 32 h at 25°C in the dark. A clear filtrate was obtained by passing the soil suspension through a 0.2-µm pore-size filter membrane (DISMIC, Advantec Toyo Kaisha, Tokyo, Japan). A 2-mL portion of the filtrate was subjected to the equilibrium pH measurement using a pH meter (HM-20E, TOA Electronics) with a glass electrode (GS-5017C, combination type for test tube, 8 mm {varphi}, TOA Electronics). A 1-mL portion of the filtrate was subjected to the determination of 2,4-D concentration using a high performance liquid chromatography unit (HPLC: pump, L-6200; UV-VIS detector, L-4200H; auto-sampler, AS-2000; column oven, L-5025; chromatogram integrator, D-2500; Hitachi) equipped with a reversed-phase analytical column (Inertsil ODS-3, 5 µm, 4.6-mm i.d., 250-mm length; GL Sciences, Tokyo, Japan). In the HPLC analysis, 2,4-D was eluted with a mixed solution of 10 mL L–1 acetic acid in H2O and methanol (3:7, volume/volume) with a flow rate of 1 mL min–1, column temperature of 40°C, detected at 210 nm, and retention time of 6.1 min. There were no replications in these experiments.

Preparation of Soil Humic Acid
A humic acid was prepared according to the procedure of Yonebayashi (1988) and Yonebayashi and Hattori (1988), as follows. Humic acid was extracted from untreated UG5 soil with 0.1 M NaOH overnight at 60°C. The extract was centrifuged (20 000 x g), and the supernatant was acidified (pH 1.2) with 4 M HCl, allowed to stand overnight to separate humic acid (precipitate) from fulvic acid (supernatant), and then centrifuged. The precipitated portion (humic acid) was again dissolved by adding NaOH, centrifuged to remove soil minerals, precipitated by adding HCl, and again centrifuged to remove fulvic acid. This procedure was repeated until the acidified supernatant became light yellowish in color. The purified humic acid fraction was redissolved into a small amount of NaOH solution and centrifuged (20 000 x g) for 2 h to remove coarse minerals. After acidification (pH 1.2) of the humic acid fraction, 0.3 M HF in 0.1 M HClO4 was added and stirred for 5 h at room temperature to dissolve and remove fine minerals. Precipitated (humic acid) fraction was washed with distilled water, redissolved into a small amount of NaOH solution, dialyzed (critical MW 8000) until the electrical conductivity of the equilibrated outer solution reached 10 µS cm–1 or less, passed through an Amberlite IR-120 resin (H+ form) column, and freeze-dried (H+ humic acid). The ash content of the humic acid as determined by dry combustion (550°C) was 0.64% (Hiradate and Yamaguchi, 2003). Acidity of the humic acid, which originated primarily from its carboxylic group, was estimated by neutralization of the humic acid (H+ form) to pH 7 (COOH acidity, 4.07 molc kg–1; Hiradate and Yamaguchi, 2003). A solid-state cross-polarization magic-angle-spinning (CPMAS) 13C nuclear magnetic resonance spectrum of the H+ humic acid showed it to be rich in aromatic and carboxylic C (Hiradate and Yamaguchi, 2003; Hiradate et al., 2004).

Adsorption of 2,4-Dichlorophenoxyacetic Acid by Metal-Humate Complexes
The H+ humic acid was dissolved into a diluted NaOH solution to obtain a H+ humic acid concentration of 4.0 g L–1 and pH value of 6.0 (stock solution). A 0.5-mL portion of this stock solution, deionized water (1.5 to 4.2 mL), 1.0 M CaCl2 (0.05 or 0.5 mL), and 0.1 M HCl or NaOH (0 to 2.30 mL) were mixed in a glass centrifuge tube (12 mL) and allowed to stand overnight, and then 489 µM 2,4-D aqueous solution (0.232 mL) was added to the tube by using the continuously adjustable air displacement pipette to achieve a final volume of 5.0 mL, an amount of H+ humic acid of 2.0 mg, an equilibrium pH value between 4 and 8, a CaCl2 concentration of 0.01 or 0.1 M, and an initial 2,4-D concentration of 22.6 µM (5.0 mg L–1). In this solution, the humic acid precipitated with Ca by forming a Ca-humate complex and then the Ca-humate complex was reacted with 2,4-D. The mixture of the Ca-humate complex and 2,4-D solution was shaken at 120 rpm for 4 h at 25°C in the dark. A clear filtrate was obtained by passing the Ca-humate complex suspension through a 0.2-µm pore-size filter membrane (DISMIC, Advantec Toyo Kaisha). The equilibrium pH value and 2,4-D concentration in the filtrate were determined by a pH meter and HPLC as described earlier. The amount of 2,4-D adsorbed by the Ca-humate complex was calculated as the difference between the amount initially present in the solution and that in the filtrate. To clarify the effects of other metals on 2,4-D adsorption, similar experiments were conducted in the presence of 0.01 M AlCl3 or 0.01 M FeCl3 instead of CaCl2. The amount of 2,4-D adsorption by Al hydroxide, which was prepared by neutralization of 5.0 mL of 0.01 M AlCl3 by 0.1 M NaOH, was also compared with that by an Al-humate complex, which was prepared by mixing 5.0 mL of 0.01 M AlCl3 and 2.0 mg of H+ humic acid at an equilibrium pH value between 4 and 8. There were no replications in these experiments.


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
CONCLUSIONS
 REFERENCES
 
Chemical Properties of Soil Samples
Untreated UG5 soil showed acidic soil pH values: pH(H2O) 5.8 and pH(KCl) 4.5. The high pH(NaF) value of 11.2 and the high P retention value of 92% (that is, 92% of added P was adsorbed) for the untreated UG5 soil indicate a considerable presence of active surface hydroxyls, which can act as ion exchange sites (Fig. 2A) and as ligand exchange sites (Fig. 2B and C). Hiradate and Uchida (2004) reported that the clay fraction (<2 µm) of the untreated UG5 soil was composed of allophane, imogolite, ferrihydrite, goethite, kaolinite, and hydroxyaluminum-vermiculite complex. These clay components, together with Al- and Fe-SOM complexes, are the major sources of the active surface hydroxyls.

Almost all organic C in the untreated UG5 soil was removed by H2O2 treatment (UG5-SOM, Table 1). Aluminum and Fe complexed with SOM in the untreated UG5 soil were probably released and transformed into Al and Fe hydroxides as newly formed solids (precipitates) in the UG5-SOM sample (Hiradate and Uchida, 2004), although some Al and Fe were lost during the preparation procedure. In UG5-SOM-AM a part of ferrihydrite, allophane, and imogolite was removed, as indicated by the reduced AlOX, FeOX, and SiOX values. In UG5-SOM-FM, crystalline iron (hydr)oxides in addition to ferrihydrite, allophane, and imogolite were completely removed, as indicated by the very low AlOX, FeOX, SiOX, and FeDCB values (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. Characteristics of untreated and chemically treated soil samples used in the present study.

 
Kinetics of 2,4-Dichlorophenoxyacetic Acid Adsorption on Untreated UG5 Soil
The standard application rate of 2,4-D on an agricultural field is 0.28 to 2.3 kg ha–1 (Tomlin, 2003), resulting in 0.56 to 4.6 mg L–1 (2.53 to 20.8 µM) of maximum 2,4-D concentration in paddy-flooded water if the flood depth is assumed to be 5 cm. In upland fields, 300 to 2500 mg L–1 of 2,4-D is the usual spraying concentration used but the concentration of 2,4-D likely to reach the soil surface could be extremely diluted by emergent plants and after rainfall. In the present study, therefore, the concentration of 2,4-D being reacted with the soil samples was set at 5.0 mg L–1 (2.26 µM), which could be comparable to field conditions. This concentration has also been recommended in the Organization for Economic Cooperation and Development (OECD) test guidelines for pesticide adsorption/desorption (OECD, 1981).

When a 2,4-D aqueous solution was mixed with untreated UG5 soil, the concentration of soluble 2,4-D decreased during the reaction period (Fig. 3). The decrease in 2,4-D concentration was faster at lower equilibrium pH values between 4.5 and 6.5; most of the decline in 2,4-D concentration occurred in the first several hours. The rapid decrease in 2,4-D concentration in solution could be caused by the adsorption of 2,4-D by the soil. Microbial degradation has also been reported to be partly responsible for 2,4-D disappearance in soil-solution systems, but microbial degradation will only be observable after several days of reaction time (Miwa and Kuwatsuka, 1991; Estrella et al., 1993; Johnson et al., 1995; Mallawatantri et al., 1996). Thus, in this experiment, it was assumed that the rapid decrease in 2,4-D concentration in solution which occurred during the first 4 h was due to adsorption reactions. The amount of 2,4-D adsorption was then compared among untreated and chemically treated soil samples to identify the important soil components involved in the adsorption of 2,4-D by the soil.


Figure 3
View larger version (20K):
[in this window]
[in a new window]

  		Fig. 3. Kinetics of 2,4-D concentration changes in aqueous solution in the presence of an Andosol (untreated UG5 soil) at equilibrium pH 4.5, 5.0, 5.5, 6.0, and 6.5. Initial 2,4-D concentration, 22.6 µM (5.0 mg L–1); the amount of soil, 50 g L–1 (oven-dry basis); background electrolyte, 0.01 M CaCl2. First, a relationship between 2,4-D concentration in aqueous solution in the presence of untreated UG5 soil and equilibrium pH was drawn at each reaction time (6 to 7 plots between pH 4 and 7), and then 2,4-D concentration at equilibrium pH 4.5, 5.0, 5.5, 6.0, and 6.5 at each reaction time was read from the relationship and plotted on this figure. The equilibrium pH value increased 0 to 0.5 units with reaction time depending on the initial pH value.

 
Soil Components Adsorbing 2,4-Dichlorophenoxyacetic Acid
The amount of 2,4-D adsorption increased with decreasing equilibrium pH in the untreated UG5 soil (Fig. 4, •). A similar trend was reported for 2,4-D adsorption by iron (hydr)oxides (Watson et al., 1973; Kavanagh et al., 1977), sandy and loamy soils in Belgium (Moreale and van Bladel, 1980), and Oxisols (Barriuso et al., 1992). Although untreated UG5 soil contained a large amount of organic C (71.9 g kg–1), removal of SOM had little effect on the amount of 2,4-D adsorbed (Fig. 4, • and {circ}). This indicates that SOM was not a major 2,4-D adsorbent, and that hydrophobic interactions may not be a major adsorption mechanism. On the other hand, removal of the SOM plus active and free metal (hydr)oxides (Fig. 4, {square} and {diamond}) decreased the amount of 2,4-D adsorption markedly, suggesting that the active surface hydroxyls derived from the active and free metal (hydr)oxides, such as allophane, imogolite, ferrihydrite, goethite, and metal-SOM complexes might have an important role in the adsorption process. The hydroxyaluminum-vermiculite complex and kaolinite would not be important for 2,4-D adsorption by untreated UG5 soil because the amount of 2,4-D adsorption on UG5-SOM-FM was very small (hydroxyaluminum-vermiculite complex and kaolinite would remain in UG5-SOM-FM).


Figure 4
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 4. The amount of 2,4-D adsorbed on untreated UG5 soil (•), UG5-SOM ({circ}), UG5-SOM-AM ({square}), and UG5-SOM-FM ({diamond}) in the presence of 0.01 M CaCl2 as a function of equilibrium pH. Reaction time, 4 h; initial 2,4-D concentration, 22.6 µM (5.0 mg L–1); the amount of soil, 50 g L–1 (oven-dried and untreated soil basis). Had the soil adsorbed all 2,4-D in this experiment, the amount of 2,4-D adsorption would have reached 0.452 mmol kg–1.

 
Mechanism of 2,4-Dichlorophenoxyacetic Acid Adsorption Caused by Active Surface Hydroxyls
The active surface hydroxyls could contribute to 2,4-D adsorption via two possible mechanisms: an ion exchange reaction and/or a ligand exchange reaction. In an ion exchange reaction, positively charged active surface hydroxyls may attract negatively charged 2,4-D through electrostatic interaction (Fig. 2A). Because the pKa value of 2,4-D is 2.81 (Bintein and Devillers, 1994), almost all 2,4-D molecules are likely to dissociate and develop a charge of –1 in the equilibrium pH range between 4 and 8. The active surface hydroxyls are amphoteric and a net positive charge develops at equilibrium pH values less than the point of zero charge. In typical allophanic Andosols, the anion exchange capacity of a SOM-rich surface horizon is almost zero and that of a SOM-poor subsurface horizon is in the range of 0 to 150 mmol kg–1 at equilibrium pH 4 to 8 (Wada, 1986). Figure 4 indicates that the amount of 2,4-D adsorbed by the SOM-rich untreated UG5 soil (•) reached 0.2 mmol kg–1 at equilibrium pH 4.5. This experiment was conducted in the presence of 0.01 M CaCl2, indicating that the initial concentration of chloride ion (Cl–) in the solution was almost 1000 times higher than that of 2,4-D anions. Therefore, 2,4-D was selectively adsorbed over Cl– on untreated UG5 soil. The selective adsorption of 2,4-D is also clearly shown in Fig. 5, in which 2,4-D adsorption on untreated UG5 soil in the presence of 0.01 M CaCl2 was essentially the same as that in the presence of 0.1 M CaCl2. The highly selective 2,4-D adsorption characteristics could not be explained by an ion exchange reaction but it could be explained by a ligand exchange reaction.


Figure 5
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 5. Effects of background electrolyte concentrations (•, 0.01 M CaCl2; {circ}, 0.1 M CaCl2) on the amount of 2,4-D adsorption on untreated UG5 soil as a function of equilibrium pH. Reaction time, 4 h; initial 2,4-D concentration, 22.6 µM (5.0 mg L–1); the amount of soil, 50 g L–1 (oven-dry basis). Had the soil adsorbed all 2,4-D in this experiment, the amount of 2,4-D adsorption would have reached 0.452 mmol kg–1.

 
It is well known that the active surface hydroxyls can adsorb organic acids and phosphate by ligand exchange reactions (Deng and Dixon, 2002). In general, the adsorption affinity of organic acids on the surface of metal (hydr)oxides is much less than that of phosphate. For example, mugineic acid, which is an Fe3+–chelating tri-carboxylic acid, is adsorbed by iron (hydr)oxides with much lower affinity than phosphate (Hiradate and Inoue, 1998). Hiradate and Uchida (2004) studied the adsorption behavior of phosphate on the same series of soil samples (untreated UG5, UG5-SOM, UG5-SOM-AM, and UG5-SOM-FM) as a function of equilibrium pH, and the amount of adsorption of 2,4-D on UG5 soil (Fig. 6A) was almost 1/100 that of phosphate (Fig. 6B). This explains competitive adsorption of phosphate over 2,4-D reported by Madrid and Diaz-Barrientos (1991): presence of phosphate (<10 mM) drastically reduced 2,4-D adsorption by an iron oxyhydroxide, lepidocrocite. Therefore, adsorption affinity of ions on the free and active metal (hydr)oxides was in the following order: phosphate>2,4-D>Cl–. The affinity sequence can be reasonably explained by a ligand exchange reaction which replaced the active surface hydroxyls on metals with the carboxyl group of 2,4-D forming a coordination bond (Fig. 2B and C). The coordination reaction would result in irreversible and hysteretic adsorption, reducing desorption, and reducing bioavailability. Furubayashi et al. (2005) reported that the dose of 2,4-D required to give 50% root inhibition of lettuce (Lactuca sativa) grown on agar medium in the presence of untreated UG5 soil was almost three times larger than that in the absence of the untreated UG5 soil (agar medium only).


Figure 6
View larger version (31K):
[in this window]
[in a new window]

  		Fig. 6. Comparison of adsorption isotherms of (A) 2,4-D and those of (B) phosphate on untreated UG5 soil. (A) The amount of soil, 50 g L–1 (oven-dry basis); background electrolyte, 0.01 M CaCl2; reaction time, 4 h. First, a relationship between 2,4-D adsorption and equilibrium pH was drawn at each initial 2,4-D concentration (6 to 7 plots between pH 3.5 and 7.5), and then 2,4-D adsorption at pH 4.0, 5.0, 6.0, and 7.0 at each initial 2,4-D concentration was read from the relationship and plotted on this figure. (B) The amounts of phosphate adsorption on untreated UG5 soil were calculated based on Hiradate and Uchida (2004).

 
Role of Soil Organic Matter in 2,4-Dichlorophenoxyacetic Acid Adsorption
Although it was clarified that the active surface hydroxyls were the dominant 2,4-D adsorbents in untreated UG5 soil, the role of SOM on 2,4-D adsorption yet remains unclear because SOM has been frequently reported to enhance adsorption of 2,4-D (Moreale and van Bladel, 1980; Mallawatantri and Mulla, 1992; Benoit et al., 1996; Cox et al., 2001). To clarify the role of SOM in 2,4-D adsorption, the adsorption characteristics of a purified humic acid extracted from untreated UG5 soil were investigated. In general, Japanese Andosols accumulate a large amount of SOM in their thick surface horizons (usually up to 50 g C kg–1; 150 g C kg–1 in extreme cases; Wada, 1986; Shoji et al., 1993) as a form of Al- and Fe-SOM complexes and more than half of the SOM is typically composed of humic acid (Wada, 1986). Therefore, we assumed that the prepared humic acid was representative of the SOM of the untreated UG5 soil. To exclude the influences of Al and Fe on 2,4-D adsorption by the humic acid, we thoroughly purified the humic acid by saturating the cation exchange sites on the humic acid with H+. Because the H+ type of humic acid prepared in the present study had an extremely low ash content (0.64%), it was water-soluble in weakly acidic conditions and required polycations (e.g., Ca2+, Al3+, Fe3+) for it to be separated as a precipitate (metal-humate complex). Thus, interactions between humic acid and 2,4-D were investigated in the presence of excess polycations so that the negative charge on the carboxylic group of SOM would be canceled by the polycations.

Figure 7A showed that the Ca-humate complex (• and {circ}) removed only a small amount of 2,4-D from the solution, suggesting that the purified humic acid did not contain Al and Fe active for 2,4-D adsorption and that any hydrophobic interactions between the humic acid and 2,4-D were very weak, especially in the higher equilibrium pH range. On the other hand, a significant amount of 2,4-D was removed from solution by the Al- ({blacksquare}) and Fe-humate complexes ({blacktriangleup}). This observation favors the importance of active surface hydroxyls on humic acid-complexed Al and Fe for 2,4-D adsorption. In this experimental system, 8.14 µmol of COOH groups on the humic acid and 50 µmol of Al or Fe were present. Therefore, a part of Al and Fe could not react with humic acid, and precipitates of Al and Fe hydroxides would be formed. It should be noted, however, that 2,4-D adsorption in the presence of 50 µmol of Al was significantly increased by the addition of the humic acid (Fig. 7B). This might be because the humic acid inhibited the hydrolysis and polymerization of Al (and Fe) by a complexation reaction and adsorption sites on Al (and Fe) could be kept available for 2,4-D adsorption (Fig. 2C). Similar effects of SOM on phosphate sorption have been reported by Gerke (1993) and Hiradate and Uchida (2004). In the case of the Al-humate complex, the maximum adsorption of 2,4-D was observed at an equilibrium pH value of about 4.3. At an equilibrium pH > 4.3, Al could be present as Al-humate complex and Al hydroxides, and the active surface hydroxyls on Al could act as an exchangeable ligand for 2,4-D adsorption. At equilibrium pH < 4.3, the Al-humate complex and Al hydroxides might be dissolved, and they would release Al as an ion, resulting in a decrease in the number of 2,4-D adsorption sites. In the case of the Fe-humate complex, 2,4-D adsorption increased as equilibrium pH declined. We hypothesize that little Fe3+ was released from the Fe-humate complex and that due to its high stability dissolution of Fe hydroxide did not occur in this equilibrium pH range.


Figure 7
View larger version (25K):
[in this window]
[in a new window]

  		Fig. 7. Comparison of 2,4-D adsorption ability among metal-humate complexes. (A) humic acid (HA) was complexed in the presence of 0.01 M CaCl2 (•), 0.1 M CaCl2 ({circ}), 0.01 M FeCl3 ({blacktriangleup}), and 0.01 M AlCl3 ({blacksquare}). (B) Comparison between Al-humate complex (humic acid was complexed in the presence of 0.01 M AlCl3, {blacksquare}) and Al hydroxides (0.01 M AlCl3 was precipitated by adding NaOH in the absence of HA, {square}). Reaction time, 4 h; initial 2,4-D concentration, 22.6 µM (5.0 mg L–1); humic acid added, 2.0 mg; COOH added, 8.14 µmol; Al or Fe added, 50 µmol; Ca added, 50 or 500 µmol.

 
In the model described above, 2,4-D molecules were adsorbed by the UG5 Andosol by ligand exchange reactions in which active surface hydroxyls associated with metals were replaced with the carboxyl group of 2,4-D. Under the conditions of our experiments, neither SOM nor humic acid adsorbed 2,4-D directly. We hypothesize that in the presence of Fe or Al hydroxides humic acid increased 2,4-D adsorption by preventing polymerization of metal hydroxides, and thus preserving more active surface hydroxyls that could adsorb 2,4-D through complexation reactions. It is possible that fulvic acid and humin fractions of SOM could act similarly because they also contain carboxylic acid groups.

Barriuso and Calvet (1992) examined 2,4-D adsorption on 54 soils and found that the amount of 2,4-D adsorption was not correlated with SOM content and occurred in the following order: Andosols>Ferralsols>>Cambisols, Rendzinas, and Vertisols. This result could be well explained by 2,4-D adsorption by active surface hydroxyls and by the effect of SOM on 2,4-D adsorption postulated in the present study: Andosols are typically acidic and rich in active surface hydroxyls and SOM; Ferralsols are acidic and relatively rich in the active surface hydroxyls but poor in SOM; Cambisols, Rendzinas, and Vertisols are weakly acidic to weakly alkaline and relatively poor in active surface hydroxyls and SOM. Some studies (e.g., Benoit et al., 1996; Cox et al., 2001) have reported 2,4-D adsorption by SOM, but attention should be paid to whether the SOM contained Al and Fe and to whether the effects of SOM addition to soils on 2,4-D adsorption were evaluated at the same equilibrium pH value.

Adsorption Behaviors of Other Organic Acids as Referenced by 2,4-Dichlorophenoxyacetic Acid
The herbicide 2,4-D has been used as a reference compound in testing adsorption/desorption behavior of chemicals in soils in many guidelines (e.g., OECD, 1981). Therefore, information on 2,4-D adsorption characteristics would be available in many soils, and this information would be a good index for estimating the behavior of a similar class of chemicals, that is, organic acids. As in the case of 2,4-D, other organic acids may also have three possible adsorption mechanisms on soils: (i) anion exchange reactions, (ii) ligand exchange reactions, and (iii) hydrophobic interactions, depending on the chemical characteristics of the organic acids, soil properties, and adsorption conditions. Most of the anion exchange reactions of soils are derived from the active surface hydroxyls associated with Al and Fe (on hydroxide surfaces, at edges of layer silicates, or complexed with organic matter). But active surface hydroxyls also adsorb organic acids by ligand exchange reactions with an affinity greater than that of anion exchange reactions. Therefore, in soils where a large amount of active surface hydroxyls are present, many organic acids may be preferentially adsorbed by ligand exchange reactions. This could be especially true for many polycarboxylic acids, because, in general, the affinity of organic acids to form complexes with Al and Fe increases exponentially with increasing numbers of carboxylic group in a molecule (e.g., formation constants (logK) for Fe3+–ligand (1:1) complexes are 4.0, 9.3, and 13.5 for acetate (monocarboxylic acid), malonate (dicarboxylic acid), and citrate (tricarboxylic acid) complexes, respectively; Stumm and Morgan, 1996). Even in the case of 2,4-D, a monocarboxylic acid, a ligand exchange reaction regulates its adsorption on an Andosol. Ligand exchange reactions are likely to play an important role in the adsorption of many natural organic acids (with weak hydrophobicity) by the soil. It is possible that the degree of adsorption of a natural organic acid on a series of Andosols would correlate with that of 2,4-D. For some synthetic organic acids with high hydrophobicity, hydrophobic interactions may need to be taken into account to predict their adsorption reactions.


   CONCLUSIONS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
REFERENCES
 
At an equilibrium 2,4-D concentration <5 mg L–1, adsorption of 2,4-D by an Andosol was interpreted to be regulated by a ligand exchange reaction in which the active surface hydroxyls on Al and Fe were replaced by the carboxylic group of 2,4-D. This reaction could account for the selective adsorption of 2,4-D over Cl–, competitive adsorption of phosphate over 2,4-D, reduction in the plant-growth-inhibitory activity of soil-adsorbed 2,4-D, and high 2,4-D adsorption ability of Andosols. The presence of SOM could enhance 2,4-D adsorption by inhibiting the hydrolysis and polymerization of Al and Fe hydroxides by complexation reactions, resulting in preservation of more adsorption sites on the hydroxide surfaces that are available for 2,4-D adsorption. The adsorption behavior of 2,4-D in soils could also be a good index for predicting the adsorption behavior of other organic acids in soils.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 CONCLUSIONS
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Hiradate, S.
Right arrow Articles by Fujii, Y.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Hiradate, S.
Right arrow Articles by Fujii, Y.
Agricola
Right arrow Articles by Hiradate, S.
Right arrow Articles by Fujii, Y.
Related Collections
Right arrow Agricultural Pesticides
Right arrow Soil Organic Matter
Right arrow Soil Surface Chemistry
Right arrow Sorption/Exchange
Right arrow Volcanic Soils

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Crop Science
Journal of Natural Resources
and Life Sciences Education		 Vadose Zone Journal
Soil Science Society of America Journal Journal of Plant Registrations The Plant Genome